Another baseball season is upon us, which means it’s time to look at the tea leaves (the projection systems) and see where our favorite teams are likely to finish. However, projection systems only take us so far. The gap between projection and reality is impacted each year by a whole host of volatile elements. One of the biggest such elements is bullpen performance.
Bullpens are notoriously volatile. Projections are built on small sample sizes- pitchers amassing data sets as small as 35 IP for LOOGYs, 90 IP at most for any reliever. Moreover, many relief innings come from minor leaguers making their debuts- pitchers who have yet to establish a major league baseline for performance.
Compounding all of this in the world of projections is the fact that relief innings are frequently the most high leverage in games. As such, they’re capable of creating a lot of noise in the gap between projections and reality. In recent years, the 2007 Diamondbacks, 2012 Orioles, and 2014 Royals and Mariners employed impressive bullpens that helped them significantly overcome both their pre-season projections and their pythagorean winning percentage.
Keeping all of this in mind, it’s fair to ask- which teams have projections that are most reliant on bullpen production? Or put another way, how volatile are MLB teams based on their bullpen this season? Let’s take a look. I’ve taken each team’s projected bullpen WAR and divided it by projected total team WAR, using the data from FanGraphs’ Depth Charts. This should tell us how much each team needs their bullpen to perform as expected to reach their projected team won-loss.
As always, click on the image to enlarge.